Modelling leaf mass per area in forest canopy as affected by prevailing radiation conditions

Leaf mass per area (LMA) is a key leaf trait, which conditions the assessment of carbon balance and the adaptation of the species to their environment. LMA decreases exponentially within the canopy at a lower rate coefficient (kLMA) than the extinction coefficient (kPAR) of photosynthetically active radiation (PAR); consequently the canopy is not fully optimized for the carbon balance. A new algorithm to simulate LMA in forest canopies is developed. The algorithm is based on a relationship between LMA of leaves at a given canopy depth and PAR, that they absorb, during leaf growth. The LMA sub-model is then tested against independent experimental data to demonstrate its validity to assess (i) the LMA vertical distribution inside the canopy, (ii) its evolution during the season and (iii) the variability observed between sites, years and species. Then, the LMA sub-model is coupled with a process-based model simulating carbon, water and energy balances in forest ecosystem. The coupled model is applied to a sensitive analysis for a case study in a beech forest.